Search

During the last decade, the use of recommender systems has been increasingly growing to the point that, nowadays, the success of many well-known services depends on these technologies. Recommenders Systems help people to tackle the choice overload problem by effectively presenting new content adapted to the user¿s preferences. However, current recommendation algorithms commonly suffer from data sparsity, which refers to the incapability of producing acceptable recommendations until a minimum amount of users¿ ratings are available for training the prediction models. This thesis investigates how the distributional semantics of concepts describing the entities of the recommendation space can be exploited to mitigate the data-sparsity problem and improve the prediction accuracy with respect to state-of-the-art recommendation techniques. The fundamental idea behind distributional semantics is that concepts repeatedly co-occurring in the same context or usage tend to be related. In this thesis, we propose and evaluate two novel semantically-enhanced prediction models that address the sparsity-related limitations: (1) a content-based approach, which exploits the distributional semantics of item¿s attributes during item and user-profile matching, and (2) a context-aware recommendation approach that exploits the distributional semantics of contextual conditions during context modeling. We demonstrate in an exhaustive experimental evaluation that the proposed algorithms outperform state-of-the-art ones, especially when data are sparse. Finally, this thesis presents a recommendation framework, which extends the widespread machine learning library Apache Mahout, including all the proposed and evaluated recommendation algorithms as well as a tool for offline evaluation and meta-parameter optimization. The framework has been developed to allow other researchers to reproduce the described evaluation experiments and make new progress on the Recommender Systems field easier, Durant l'última dècada, l'ús dels sistemes de recomanació s'ha vist incrementat fins al punt que, actualment, l'èxit de molts dels serveis web més coneguts depèn en aquesta tecnologia. Els Sistemes de Recomanació ajuden als usuaris a trobar els productes o serveis que més s¿adeqüen als seus interessos i preferències. Una gran limitació dels algoritmes de recomanació actuals és el problema de "data-sparsity", que es refereix a la incapacitat d'aquests sistemes de generar recomanacions precises fins que un cert nombre de votacions d'usuari és disponible per entrenar els models de predicció. Per mitigar aquest problema i millorar així la precisió de predicció de les tècniques de recomanació que conformen l'estat de l'art, en aquesta tesi hem investigat diferents maneres d'aprofitar la semàntica distribucional dels conceptes que descriuen les entitats que conformen l'espai del problema de la recomanació, principalment, els objectes a recomanar i la informació contextual. En la semàntica distribucional s'assumeix la següent hipotesi: conceptes que coincideixen repetidament en el mateix context o ús tendeixen a estar semànticament relacionats. Concretament, en aquesta tesi hem proposat i avaluat dos algoritmes de recomanació que fan ús de la semàntica distribucional per mitigar el problem de "data-sparsity": (1) un model basat en contingut que explota les similituds distribucionals dels atributs que representen els objectes a recomanar durant el càlcul de la correspondència entre els perfils d'usuari i dels objectes; (2) un model de recomanació contextual que fa ús de les similituds distribucionals entre condicions contextuals durant la representació del context. Mitjançant una avaluació experimental exhaustiva dels models de recomanació proposats hem demostrat la seva efectivitat en situacions de falta de dades, confirmant que poden millorar la precisió d'algoritmes que conformen l'estat de l'art. Finalment, aquesta tesi presenta una llibreria pel desenvolupament i avalu, Postprint (published version)

Several research works have demonstrated that if users' ratings are truly context-dependent, then Context-Aware Recommender Systems can outperform traditional recommenders. In this paper we present a novel contextual pre-filtering approach that exploits the implicit semantic similarity of contextual situations. For determining such a similarity we rely only on the available users' ratings and we deem as similar two syntactically different contextual situations that are actually influencing in a similar way the user's rating behavior. We validate the proposed approach using two contextually tagged ratings data sets showing that it outperforms a traditional pre-filtering approach and a state-of-the-art context-aware Matrix Factorization model., Peer Reviewed, Postprint (published version)

Environmental problems should not be tackled by scientists or policy makers alone. Involving the general public in observing and understanding our changing world, and encouraging citizen stewardship for the (marine) environment are crucial elements for a sustainable way of facing current and future problems. The EC-funded project Citclops will start in October 2012 and is focused on retrieving, through crowdsourcing, data on three main optical properties related to sea-water quality: colour, transparency and fluorescence. Novel technologies will be developed to retrieve these properties based on citizens’ measurements with common mobile devices. Concerning water colour, smart-phone camera images, taken through citizens’ collaboration, will be used to calculate the Forel-Ule index for the water body, providing an indication of gross biological activity. Water-transparency is planned to be measured by three alternative sensor systems : i) smart-phone cameras; ii) low-cost moored instrumentation; and iii) underwater, wearable cameras with added low-cost sensing systems (quasi-digital sensors). Obtained ocean colour and transparency data will complement the longterm global data series which go back to the late 20th century. Finally, to assess the fluorescence of different water constituents, low-cost sensors and light-sources will be customized and connected to smart-phones and other mobile devices. All acquired data is automatically uploaded through a specific service or application (such as Google+ Instant Upload), archived remotely and processed, and resulting information is accessed through a webpage or a mobile application by end users, thus closing the loop to citizens and policy makers

Web services have attracted attention as a possible solution to share knowledge and application logic among different heterogeneous agents. A classic approach to this subject is using SOAP, a W3C protocol aimed to exchange structured information. The Web Services Interoperability organization (WS-I), defines a set of extensions, commonly called WS-*, which further enhance this knowledge exchange defining mechanisms and functionalities such as security, addressability or service composition. This thesis explores a relatively new alternative approach to the SOAP/WS-I stack: REST-based Web services. The acronym REST stands for Representational state transfer; this basically means that each unique URL is a representation of some object. You can get the contents of that object using an HTTP GET; you then might use a POST, PUT or DELETE to modify the object (in practice most of the services use a POST for this). All of Yahoo’s Web services use REST, including Flickr; del.icio.us API uses it; pubsub [http://www.pubsub.com/], Bloglines [http://www.bloglines.com/], Technorati [http://technorati.com/] and both, eBay and Amazon, have Web services for both REST and SOAP. Google seems to be consistent in implementing their Web services to use SOAP, with the exception of Blogger, which uses XML-RPC. The companies and organization that are using REST APIs have not been around for very long, and their APIs came out in the last seven years mostly. So REST is a new way to create and integrate Web services, whose main advantages are: being lightweight (not a lot of extra xml mark-up), human readable results, easy to build services (no toolkits required). Although REST is still generating discussion about possible implementations, and different proposals have been put forward, it provides enough mechanisms to allow knowledge-representations sharing among heterogeneous intelligent services. In this thesis, a novel way to integrate intelligent Web-services is designed and developed, and

This paper presents an agent-based methodological approach to design distributed service-oriented systems which can adapt their behaviour according to changes in the environment and in the user needs, even taking the initiative to make suggestions and proactive choices. The highly dynamic, regulated, complex nature of the distributed, interconnected services is tackled through a methodological framework composed of three interconnected levels. The framework relies on coordination and organisational techniques, as well as on semantically annotated Web services to design, deploy and maintain a distributed system, using both a top-down and bottom-up approach. We present results based on a real use case: interactive community displays with tourist information and services, dynamically personalised according to user context and preferences., Preprint

Personalization and recommendation systems are a solution to the problem of content overload, especially in large information systems. In this thesis, a personalized recommendation system enhanced with semantic knowledge has been developed in order to overcome the most common limitations of traditional approaches: the cold-start and the sparsity problems. The recommender consists of the following two main components. A user-profile learning algorithm combines user’s feedback from different channels and employs domain inferences to construct accurate user profiles. A recommendation algorithm, using content-based filtering, exploits the semantic structure of the domain to obtain accurate predictions and generate the corresponding recommendations. The system’s design proposed is flexible enough to be potentially applied to applications of any domain that can be properly described using ontologies. In addition to the development of the recommendation system, an existing Web-application in the tourism domain has been extended and adapted in order to be able to integrate the recommender into it. The overall recommendation system has been evaluated and the results obtained indicate that it satisfies the requirements established.

Interconnected service providers constitute a highly dynamic, complex, distributed environment. Multi-agent system design-methodologies have been try-ing to address this kind of environments for a long time. The European project ALIVE presents a framework of three interconnected levels that tackles this issue relying on organisation and coordination techniques, as well as on developments in the Web-services world. This paper presents initial results focused on a high-tech, real use case: interactive community displays with touristic information and services, dynamically personalized according to user preferences and local laws., Peer Reviewed, Preprint

This paper presents a case study for the development of an integration framework in the area of collaborative work environments. This framework is based on REST Web services, and has been designed and implemented as part of the Laboranova European project., Peer Reviewed, Preprint

La falta de servicios automáticos y personalizados de información fácilmente accesibles para los turistas y ciudadanos cuándo éstos se encuentran en la vía pública, impacta negativamente en el sector turístico, ya que implica una disminución del uso de los servicios que ofrece la ciudad y del grado de satisfacción de los usuarios. El proyecto SOPAT ha desarrollado un servicio personalizado de información y guía de ciudad que facilita el acceso a la información de interés a todo tipo de turistas y ciudadanos utilizando como plataforma experimental un punto de información multimedia con servicios interactivos en la zona pública., The lack of automatic, personalized services of information, which were easily accessible to tourists and citizens when they are in public spaces, negatively impacts the tourism sector, since it reduces the use of the services offered by the city and the satisfaction degree of the users. The SOPAT project has developed a personalized service of information and city guide that facilitates all kinds of tourists and citizens access to information of interest using an interactive community display as experimental platform.

Con este proyecto se quiere mejorar uno de los productos principales de la empresa, que consiste en un sistema de televisión interactiva que permite a los usuarios poder disfrutar de varios servicios tales como una aplicación ofimática, una guía local y la adquisición de contenidos de distintos tipos: películas, series, clips, música, información cultural y videojuegos entre otros. Actualmente existen varios sistemas de recomendación de contenido de vídeo que funcionan relativamente bien, pero únicamente en entornos con un gran número de usuarios que continuamente interactúan con el sistema. Esto implica, que los algoritmos de recomendación de contenido de vídeo actuales sólo tengan cabida en el entorno Web, dónde el número de usuarios puede llegar a ser considerable y se dispone de una alta densidad de información. Se pretende diseñar un sistema de recomendación en el que la precisión de las recomendaciones no dependa tanto de la densidad de información disponible, sino más bien en la calidad de la información, dado que estas son las características del entorno donde se implantará el sistema.

In this report an unsupervised and knowledge-based algorithm for concept sense disambiguation in concept maps is proposed. Concept maps are graphical tools for organizing and representing knowledge, based on concepts and labeled interconnections among them, forming propositions. The disambiguation process is carried combining Magnini’s domain, context information and the gloss. It’s supported in the Spanish WordNet lexical database and the lexical relations hypernyms-hyponyms, meronyms-holonyms and instance., Postprint (published version)

This document contains requirements and specifications for the data-storage layer of Laboranova’s integration architecture and for the data structure of the idea concept. This document presents technology choices for the implementation of the data-storage layer in a way that facilitates composing the different sub-projects (SPs) of Laboranova into a modular collaborative environment. The idea concept is central in Laboranova and working with ideas involves several aspects: • a conceptualisation of what an idea is; • data structures for ideas (Representations based on these data structures are passed among the software systems which make up the Laboranova environment.); • the integration of idea’s manipulation carried out by the different Laboranova’s tools. This document focuses on the data structures, taking into account the requirements of the visualization tool and the results obtained from a questionnaire (‘Laboranova Data Storage Questionnaire’) which was conducted in May 2007 among Laboranova’s partners. Such questionnaire was intended to gather requirements on technology and integration choices which affect some aspects of the idea data structure. This document also takes into account the user requirements gathered in D6.1.1 (‘Report on deployment scenarios and requirements’, version 1.0), which concern: • Idea management: archival and retrieval of ideas • Versioning: branching; vertical expansion of an idea; originating new ideas changing a first one • Relations between ideas: idea clustering • Flexibility and transparency: integrative design, providing a modular user environment • Integration with current practices: connecting Laboranova with previous tools and processes This document is divided in the following parts: • description of the integration architecture to frame this document with the context of Laboranova and the SPs that compose it; • overall description of the data layer; • conceptualization of the idea data structure by: A) giving a compilati, Postprint (published version)

El desarrollo y utilización de las ontologías, se está viendo cada vez más en ascenso, demostrando con ello su gran utilidad en los contextos en los que se necesite formalizar y compartir un conocimiento. Cada vez más se está requiriendo la creación de un ambiente que facilite la colaboración en el trabajo con las ontologías y hoy no están creadas las condiciones para esto. El proceso de elaboración de ontologías, las herramientas disponibles para su obtención y mantenimiento y los lenguajes de especificación que se están utilizando, tienen un alto grado de complejidad para su interpretación y amplia utilización por parte de personas no son expertas en el tema. Esto sugiere la utilización de herramientas más simple, que vinculada con las ontologías pueda ser comprendida y mejor aprovechadas por el hombre y que a su vez mantenga las capacidades para su interpretación computacionalmente. Los mapas conceptuales son una herramienta muy intuitiva para la representación del conocimiento que posee el hombre. Las facilidades que brinda en su elaboración, flexibilidad y estructura, lo han convertido en una herramienta muy útil para la gestión del conocimiento y la colaboración. La flexibilidad de los mapas conceptuales, lo convierte en una herramienta muy poco formal, a diferencia de las ontologías, sin embargo, se considera que esto no impide pensar que se puedan integrar con las ontologías y ser utilizados como un lenguaje de representación gráfico conceptual de estas. En esta investigación se propone la definición de un modelo que permite la integración entre las ontologías y los mapas conceptuales. El Modelo Unificado, como se le ha llamado, es resultado de la unificación de algunas características resultantes de análisis realizados a diferentes formas de representación del conocimiento de la Inteligencia Artificial y los mapas conceptuales. La características identificadas e integradas generó un conjunto de criterios que fundamentan la integración propuesta en ontología, Postprint (published version)

The work described in this document has been carried out in the frame of the @LIS TechNET project, which aims to create a live network across Europe and Latin America as a new, virtual, and interactive learning environment, based on software-agent technology. The purpose of this document is to describe a visualization tool developed for the TechNET's planning system. As background, this document also briefly describes what mixed-initiative planning is., Postprint (published version)

Les contribucions d'aquesta tesi uneixen dues disciplines: ciències ambientals (específicament, gestió d'aigües residuals) i informàtica (específicament, intel·ligència artificial). El tractament d'aigües residuals com a disciplina opera fent servir un rang de diferents enfocaments i mètodes que inclouen: control manual, control automàtic on-line, modelat numèric o no-numèric, models estadístics i simulacions. La tesi caracteritza la recerca interdisciplinària de tècniques d'intel·ligència artificial (raonament basat en regles, raonament basat en casos, ontologies i planificació) a sistemes de suport a la decisió a l'entorn ambiental. El disseny de l'arquitectura d'aquesta aplicació, el sistema OntoWEDSS, augmenta els sistemes clàsics de raonament existents (raonament basat en regles i basat en casos) amb una ontologia de domini per a la gestió de plantes de tractament d'aigües residuals. La integració de l'ontologia WaWO recentment creada proporciona a OntoWEDSS una major flexibilitat en la capacitat de gestió. La construcció del sistema de suport a la decisió OntoWEDSS es basa en l'estudi d'un cas específic, però el sistema també és d'interès general ja que l'arquitectura basada en l'ontologia pot aplicar-se a qualsevol estació depuradora i, a un nivell apropiat d'abstracció, a altres dominis ambientals. El sistema OntoWEDSS millora la diagnosi de l'estat de l'estació depuradora, proporciona suport a la solució de complexes problemes relacionats amb aigües residuals, i facilita el modelatge del coneixement i la seva reutilització mitjançant l'ontologia WaWO. En particular, a la investigació s'han aconseguit els següents objectius: (1) la millora del modelatge de la informació sobre processos de tractament d'aigües residuals i la clarificació de part de la confusió existent en la terminologia del domini, (2) la incorporació de coneixement microbiològic (referent al procés del tractament i modelat mitjançant una ontologia) dins del procés de raonament, (3) la c, Las contribuciones de esta tesis unen dos disciplinas: ciencias ambientales (específicamente, gestión de aguas residuales) e informática (específicamente, inteligencia artificial). El tratamiento de aguas residuales como disciplina opera utilizando un rango de diferentes enfoques y métodos que incluye: control automático on-line, modelado numérico o no-numérico, razonamiento basado en reglas, razonamiento basado en casos, soporte a la decisión y planificación. La tesis caracteriza una aplicación interdisciplinaria de técnicas de inteligencia artificial a sistemas de soporte a la decisión en el dominio ambiental. El diseño de la arquitectura de esta aplicación, el sistema OntoWEDSS, aumenta los sistemas híbridos de razonamiento ya existentes (razonamiento basado en reglas y basado en casos) con una ontología de dominio para la gestión de plantas de tratamiento de aguas residuales. La integración de la ontología WaWO, de nueva creación, proporciona a OntoWEDSS una mayor flexibilidad en la capacidad de gestión. La construcción del sistema de soporte a la decisión OntoWEDSS se basa en el estudio de un caso específico, pero el sistema resulta también es de interés general puesto que la arquitectura basada en ontologías puede aplicarse a cualquier planta de tratamiento de aguas residuales y, a un nivel apropiado de abstracción, a otros dominios ambientales. El sistema OntoWEDSS mejora la diagnosis del estado de la planta de tratamiento, proporciona soporte a la resolución de complejos problemas relacionados con aguas residuales, y facilita el modelado del conocimiento y su reutilización mediante la ontología WaWO. En particular, la investigación ha alcanzado los siguientes objetivos: (1) la mejora del modelado de la información sobre procesos de tratamiento de aguas residuales y la clarificación de parte de la confusión existente en la terminología relacionada, (2) la incorporación de conocimiento microbiológico (referente al proceso del tratamiento y modelado median, The contributions of this thesis bridge two disciplines: environmental science (specifically, wastewater management) and computer science (specifically, artificial intelligence). Wastewater management as a discipline operates using a range of different approaches and methods which include: manual control, on-line automatic control, numerical or non-numerical models, statistical models and simulation models. The thesis characterizes an interdisciplinary research on artificial intelligence techniques (rule-based reasoning, case-based reasoning, ontologies and planning) applied to environmental decision-support systems. The integrated architecture's design of this application, the OntoWEDSS system, augments classic reasoning systems (rule-based reasoning and case-based reasoning) with a domain ontology about the management of wastewater treatment plants. The integration of the newly created WaWO ontology provides a more flexible management capability to OntoWEDSS. The construction of the OntoWEDSS decision support system is based on a specific case study but the system is also of general interest, given that its ontology-underpinned architecture can be applied to any wastewater treatment plant and, at an appropriate level of abstraction, to other environmental domains. The OntoWEDSS system improves the diagnosis of the state of a treatment plant, provides support for wastewater-related complex problem-solving, and facilitates knowledge modeling and reuse by means of the WaWO ontology. The following research targets have been achieved in particular: (1) the improvement of the modeling of the information about wastewater treatment processes and the clarification of a part of the existing terminological confusion in the domain, (2) the incorporation of ontology-modeled microbiological knowledge related to the treatment process into the reasoning process, (3) the creation of a decision support system with three layers (perception, diagnosis and decision support) which, Questa tesi contribuisce alla intersezione di due discipline: le scienze ambientali (specificamente, la gestione delle acque di rifiuto) e la informatica (specificamente, la intelligenza artificiale). Nel trattamento delle acque di rifiuto come disciplina si utilizzano diversi metodi, che includono: controllo manuale, controllo automatico on-line, modelli numerici o non-numerici e simulazioni. La tesi caratterizza un'applicazione interdisciplinare di tecniche di intelligenza artificiale a sistemi di aiuto alla decisione in campo ambientale. L'architettura di questa applicazione, il sistema OntoWEDSS, amplia i sistemi di ragionamento ibrido esistenti (ragionamento basato su un sistema di regole, ragionamento basato sull'esperienza, aiuto alla decisione e pianificazione) con un'ontologia di dominio per la gestione di depuratori di acque di rifiuto. L'integrazione dell'ontologia WaWO, di nuova creazione, fornisce a OntoWEDSS una maggiore flessibilità nella sua capacità di gestione. La costruzione del sistema OntoWEDSS si basa sullo studio di un caso specifico, però il sistema risulta anche di interesse generale dato che l'architettura basata su un'ontologia può essere applicata a un qualsiasi depuratore e, considerando un adeguato livello d'astrazione, ad altri domini ambientali. Il sistema OntoWEDSS migliora la diagnosi dello stato del depuratore, fornisce aiuto alla soluzione di problemi complessi relazionati con le acque di rifiuto e facilita la modellizzazione della conoscenza e la sua riutilizzazione mediante l'ontologia WaWO. In particolare, la ricerca realizzata ha raggiunto i seguenti obiettivi: (1) il miglioramento dell'informazione sui processi di depurazione e il chiarimento di parte della confusione esistente nella terminologia relativa, (2) l'incorporazione di conoscenza microbiologica (riguardo al processo di depurazione e mediante la modellizzazione ontologica) nel processo di ragionamento, (3) la creazione di un sistema di aiuto alla decisione con, Postprint (published version)

We present an ontology (named WaWO - Waste Water Ontology), integrated into an existent environmental decision-support system (named DAI-DEPUR+), designed to model the information about a wastewater treatment process through the definition of the basic terms and relations comprising the vocabulary of the wastewater treatment area. The WaWO ontology is built following the ideas of Uschold and Grüninger and is a hierarchically structured set of terms and a set of axioms for describing the domain of wastewater treatment. It is the manifestation of a shared understanding of the wastewater domain that is agreed among a number of agents: mainly, experts in environmental and chemical engineering. This ontology incorporates some new features for the wastewater domain, such as: semi-automatic generation of knowledge bases, modeling of the cause-effect relations existing within a wastewater treatment plant, integration with predictive reasoning based on transition networks., Postprint (published version)

Presentation of an environmental decision-support system integrating a rule-based expert system, a case-based reasoner and an ontological knowledge-base. This system is able to model the information about a wastewater treatment process through the definition of the basic terms and relations comprising the vocabulary of the wastewater treatment area. Furthermore, this management system optimizes the operation of wastewater treatment by a more reliable management and making easier its self-portability., Postprint (published version)

Citizen science has expanded rapidly over the past decades. Yet, defining citizen science and its boundaries remained a challenge, and this is reflected in the literature-for example in the proliferation of typologies and definitions. There is a need for identifying areas of agreement and disagreement within the citizen science practitioners community on what should be considered as citizen science activity. This paper describes the development and results of a survey that examined this issue, through the use of vignettes-short case descriptions that describe an activity, while asking the respondents to rate the activity on a scale from 'not citizen science' (0%) to 'citizen science' (100%). The survey included 50 vignettes, of which five were developed as clear cases of not-citizen science activities, five as widely accepted citizen science activities and the others addressing 10 factors and 61 sub-factors that can lead to controversy about an activity. The survey has attracted 333 respondents, who provided over 5100 ratings. The analysis demonstrates the plurality of understanding of what citizen science is and calls for an open understanding of what activities are included in the field.

Citizen science has expanded rapidly over the past decades. Yet, defining citizen science and its boundaries remained a challenge, and this is reflected in the literature-for example in the proliferation of typologies and definitions. There is a need for identifying areas of agreement and disagreement within the citizen science practitioners community on what should be considered as citizen science activity. This paper describes the development and results of a survey that examined this issue, through the use of vignettes-short case descriptions that describe an activity, while asking the respondents to rate the activity on a scale from 'not citizen science' (0%) to 'citizen science' (100%). The survey included 50 vignettes, of which five were developed as clear cases of not-citizen science activities, five as widely accepted citizen science activities and the others addressing 10 factors and 61 sub-factors that can lead to controversy about an activity. The survey has attracted 333 respondents, who provided over 5100 ratings. The analysis demonstrates the plurality of understanding of what citizen science is and calls for an open understanding of what activities are included in the field.